1. Field of the Invention
The present invention relates to image forming apparatuses which form images on recoding media, such as printing paper, photosensitive paper, and electrostatic recording paper. More particularly, the present invention relates to an image forming apparatus in which toner images are transferred onto a rotary member, such as an intermediate transfer belt, from a plurality of image bearing members.
2. Description of the Related Art
Recently, there has been a demand for electrophotographic color image forming apparatuses, such as color printers and color copy machines, with high output image quality.
Factors which determine the output image quality include recording accuracy which typically affects the displacement of a print start position at which a process of printing an image on a recording medium, such as paper, is started, image expansion and contraction, etc., and color registration. The color registration is an index of accuracy with which toner images of different colors are superimposed, and affects the color of the output image. In particular, in an electrophotographic color image forming apparatus, when the environment changes or when the apparatus is used for a long term, a reduction in the recording accuracy and a variation in color due to color misregistration will occur due to variable factors of components included in the apparatus. As a result, the output image quality will be degraded.
In, for example, an image forming apparatus which uses an intermediate transfer belt as an endless belt, one of the causes of the above-mentioned variation is a variation in the velocity of the intermediate transfer belt.
Accordingly, a method described in, for example, Japanese Patent No. 2655603 is used. More specifically, toner patches of different colors are formed on the intermediate transfer belt, and the positions of the toner patches are detected by registration detection sensors. The timing at which the toner images of different colors are formed on the intermediate transfer belt is changed in accordance with the detection result, and thus the color misregistration can be reduced. Here, the toner patches are unfixed toner images used for detecting the color misregistration.
However, even if the color misregistration is corrected using the registration detection sensors according to the related art, the color misregistration generated when the toner images of the respective colors are transferred onto a recording medium cannot be completely eliminated in practice.
The main cause of this is that the peripheral velocity of the intermediate transfer belt in the process of detecting the positions of the toner patches on the belt with the registration detection sensors differs from the peripheral velocity of the belt in the actual image forming operation. The difference in the peripheral velocity of the intermediate transfer belt will be described below.
FIG. 11 is a diagram illustrating the state in which load is applied to an intermediate transfer belt unit in a tandem color image forming apparatus using a common intermediate transfer belt. Referring to FIG. 11, to improve the transfer efficiency, a peripheral velocity Vb of the intermediate transfer belt is generally set to be about 0.5% or less higher than a peripheral velocity Vd of photosensitive drums. In this case, when Tb is a torque necessary for moving only the belt and μF is a frictional force generated by the contact between the belt and the drums, a belt driving torque T can be calculated as follows:T=Tb+μF×4  (11)
In the above equation, μ is a coefficient of friction between the belt and the drums and F is a transfer pressure.
In contrast, referring to FIG. 12, if the drum peripheral velocity Vd is intentionally set to be higher than the belt peripheral velocity Vb, the belt driving torque T can be obtained as follows:T=Tb−μF×4  (12)
In this case, the belt driving torque T is reduced since the photosensitive drums rotate the belt.
Assuming that there are two kinds of coefficients of friction μ between the belt and the drums as described below, the torque T varies as in Equations (13) to (17) given below during the time period from when the belt in the stationary state is activated and to when the belt is stopped after the image forming operation. In this case, the state in which the load torque is applied to the belt changes as shown in FIGS. 13 to 20. Each of FIGS. 13 to 20 shows a structure including photosensitive drums 26, developing rollers 54, primary transfer rollers 52, and an intermediate transfer belt 30. With regard to the colors, Y, M, C, and Bk denote yellow, magenta, cyan, and black, respectively. The two kinds of coefficients of friction μ between the belt and the drums are a coefficient of friction μ1 applied when no toner is present between the belt and the drums and a coefficient of friction μ2 applied when toner is present between the belt and the drums.T=Tb+μ1F×4  (13) (see FIG. 13)T=Tb+(μ1×3+μ2F)  (14) (see FIG. 14)T=Tb+(μ1F×2+μ2F×2)  (15) (see FIG. 15)T=Tb+(μ1F+μ2F×3)  (16) (see FIG. 16)T=Tb+μ2F×4  (17) (see FIG. 17)
With regard to the states shown in FIGS. 18 to 20, FIG. 18 corresponds to Equation (16), FIG. 19 to Equation (15), and FIG. 20 to Equation 14. Then, the state returns to that shown in FIG. 13, which corresponds to Equation (13).
Referring to FIG. 14, when the developing roller 54Y comes into contact with the photosensitive drum 26Y, the toner on the developing roller 54Y is supplied to the photosensitive drum and adheres thereto irrespective of a latent image formed thereon, and then the toner is conveyed to a nip section between the photosensitive drum and the intermediate transfer belt. Therefore, the presence or absence of the toner in the nip section between the photosensitive drum and the intermediate transfer belt is determined depending on whether the developing roller in the developing device is in a contact state or a non-contact state, and does not depend on the toner actually used to form an image in accordance with the latent image.
With regard to the relationship between the magnitudes of the above-described coefficients of friction μ1 and μ2, μ1>μ2 is generally satisfied. The load (torque) applied to the belt is reduced when the developing device is in the contact state and is increased when the developing device is in the separated state.
In the image forming apparatus according to the related art, the belt driving torque applied in the process of detecting the toner patches on the belt with the registration detection sensors can be calculated as in Equation (17) and is maintained constant. In addition, the peripheral velocity of the belt is also constant. This is obviously different from the state in which the torque varies in the image forming operation.
The belt is driven by a belt drive transmission system including a gear train. As defined by the Hooke's law, the belt drive transmission system is elastically deformed by an amount proportional to a stress generated by a load torque applied by the belt drive transmission system. Due to the elastic deformation, the transmission speed of the drive transmission system, that is, the belt peripheral velocity, varies.
When each of the states corresponding Equations (13) to (17) changes to the next state, the belt peripheral velocity also changes. For example, the belt peripheral velocity decreases when the load torque applied to the belt increases, and the belt peripheral velocity increases when the load torque applied to the belt decreases.
In the image forming apparatus according to the related art, the process of detecting the toner patches on the belt with the registration detection sensors is performed while the belt velocity does not vary, and the correction is performed on the basis of the thus-obtained detection result. However, in the actual image forming operation, the belt peripheral velocity varies when the load torque largely varies. When the belt peripheral velocity varies, the color misregistration (displacement of transfer position) easily occurs.
The variation in the belt velocity can be eliminated by the following three methods. The first method is to eliminate the elastic deformation by increasing the rigidity of the belt drive transmission system. The second method is to eliminate the variation in the coefficient of friction μ between the belt and the drums. The third method is to form images after the state corresponding to Equation (17) is obtained.
The first method will be explained. In general, the above-described elastic deformation can be suppressed by increasing the rigidity of the belt drive transmission system. The rigidity can be increased by, for example, changing the material of gears included in the drive transmission system from resin, such as polyacetal, to metal, such as brass. The inventors of the present invention experimentally confirmed that the velocity variation can be reduced by increasing the rigidity by using metal gears. However, in such a case, the rigidity of the metal gears is excessively high and vibration of the gears which mesh with each other occurs. Therefore, there is a problem that the images will be affected by the generated vibration. In addition, since the metal gears are formed by machining processes, high costs are incurred compared to resin gears which can be formed by injection molding.
Next, the second method will be explained. Theoretically, variation in the coefficient of friction μ can be eliminated if the coefficients of friction μ1 and μ2 are equal to each other. However, in practice, surface layers of the photosensitive drums are smooth and easily adhere to the belt. Therefore, a considerably large frictional force is generated. Although the frictional force can be reduced by forming slightly irregular surfaces on the photosensitive drums and reducing the contact areas thereof, there is a risk that the image quality will be degraded in such a case. Therefore, this method is not practical. In addition, variation in the frictional force is due not only to the presence or absence of the toner but also to the attractive force caused by the transfer bias, and cannot be eliminated.
Next, the third method will be explained. The third method can be technically realized by starting and stopping a charging step, a developing step, and a transfer step, which are performed by image-forming process units and which cause the load variation, in a period other than the period in which visible images are transferred onto the intermediate transfer belt from the photosensitive drums. In such a case, high-quality images in which color misregistration is suppressed can be obtained. However, since the charging step, the developing step, and the transfer step performed by the above-mentioned units are started and stopped in a period other than the period in which the visible images are transferred onto the intermediate transfer belt from the photosensitive drums, the time required for the charging step, the developing step, etc., is increased and the yield of the apparatus is reduced. In addition, there is also a problem that the lives of the above-mentioned units will be reduced. This cannot be ignored, in particular, when intermittent recording is performed. In such a case, a user must frequently replace the above-mentioned units and high running costs are incurred.